GRM is always my first stop on the WTF train, so please get me straight on this.
2010 Forester
AC is cold for the first 10-15 minutes after start-up, then not cold and compressor quits cycling.
park long enough for everything to return to ambient, and the cycle repeats.
I haven't put the gauges on it yet.
does this sound familiar to anyone? And if yes, what's the most likely issue?
Charge and pressure sensor for compressor turn on.
I have a tractor at work that does this exact same thing. It is infuriating. We took it to Deere and they claimed there was nothing wrong. Wish I could be of more help. Following for potential solutions I can tell the Deere dealer what to do to fix our tractor.
There was a video I saw on the "south main auto" YouTube channel just the other day where they had a Forester doing pretty much the same thing. Turned out to be an issue with the AC clutch.
Check pressures. I've seen once on a Forester where the expansion valve would stick shut, or freeze shut, or something, and cause the high side pressure to skyrocket to the high side limit. Have not seen too many compressor clutch failures on Subarus, and by that I mean have not seen any at all. That doesn't mean it can't happen, of course, but compressor clutch issues tend to be a pattern failure kind of thing, where they either work well on all of them or fail on all of them. (See also: Honda CR-V, Volvo whiteblock)
This is exactly what my Saab C900 does. In my case it's a slipping clutch. As everything heats up, the clutch starts slipping. Mine is due to a weak coil, but it could also be a large gap between the plates. Easy enough to check - just look at the clutch when it's not working and see if its engaged and turning.
My mom's 05 Forester does the same thing.
Dealing with something similar on my mom's 2011 Outback. Hers will blow cold on startup at idle, then as you drive the car and it heats up more, it will blow warm at stop lights or at idle. Putting the car in Park or Neutral and revving the engine will bring cold air back.
In her case, I am 95% sure it's a slipping AC compressor clutch and my master-tech mechanic friend agrees with me. She has about 70k on the car so it's odd that it is failing IMO, but their local mechanic pressure-tested the system and re-filled it (with dye, to check for leaks) and it's not leaking at all. Not much else to be wrong with it, from what I know.
In reply to Pete. (l33t FS) :
Is it safe (in terms of exploding them, not in terms of crashing the car) to leave gauges attached while driving?
Background: AC wasn't cold on test drive, I bargained the price down accordingly, dude "put in a can of gas" before we picked it up, and it's been operating pretty normally for a couple weeks. I guess it's possible that he under-charged it, but why did it need a charge in the first place? Must be a leak somewhere, yes?
In reply to AngryCorvair (Forum Supporter) :
I don't see why not, I do it with fuel pressure gauges all the time. The hard part is being able to read the gauges while they are taped to a windshield.
More safety minded, you could drive it until the A/C stops working and check the pressure then.
In reply to AngryCorvair (Forum Supporter) :
Having made this mistake, it is possible to overcharge with the can. When the system heats up, it is over pressure and kicks off. In my 2010 Corolla, I had that problem make sysmptoms worse, but the underlying problem eventually turned out to be the variable compressor was going out.
Anyone know what the low- and high-side pressures should be on 2010 Forester?
I'm guessing it's undercharged.
In reply to AngryCorvair (Forum Supporter) :
There is rarely a hard pressure spec for any given vehicle because refrigerant pressure varies dependent on ambient temperature. Generally the spec is to have a specific amount of refrigerant in the system and that's usually on a label under the hood.
In reply to AngryCorvair (Forum Supporter) :
Toyman has a good chart about 6 post down.
https://grassrootsmotorsports.com/forum/grm/air-conditioning-leaks/183256/page1/
AngryCorvair (Forum Supporter) said:
Anyone know what the low- and high-side pressures should be on 2010 Forester?
Depends a loooot on ambient temperatures and humidity, which is really why the system needs to be pulled under a vacuum and filled by weight. But as a general rule: Low side should generally be 40psi or lower but not like below 30. Higher makes for poor cooling, lower makes for evaporator freezeup, and low side pressure is more of a diagnostic for undercharge or poor compressor/expansion valve operation. High side is what really depends on temps, on a cool day 175ish is okay, hotter makes pressures go right on up there. I'd be concerned over 225-250 if it wasn't cooling well. If both pressures are high, it might be overcharge, or it may be an airflow problem through the condensor. Another reason why step 1 is generally an evac and recharge to verify system charge before worrying about throwing parts at it.
Most likely you have a simple issue like the clutch kicking out, but pressures will tell you if it is kicking off because it is commanded to kick off.
Does your evaporator look like my jury rigged industrial unit? This one doesn't work all day...
I had the fan control module on the MPV quit a summer or two ago. AC worked okay until everthing got hot, then the compressor quit kicking in and the system blew warm air. The temp gauge for the engine never showed an abnormally high reading. Are the fans kicking in when the A/C is on?
Almost everywhere it's ambient temp with compressor off is what the high and low gauge read. Once compressor runs, 40 or so on the low and about 3-4 times for the high.
I just went through the exact same problem with my Volvo, and someone here mentioned the "Volvo Zip Tie Fix". I'll try to find the link and give him proper credit.
I did it, and three weeks later, I still have ice cold air all the time.
And it was essentially free (plus the cost of three zip ties).
I have seen this happen from an overcharge, or a hi-limit switch that is starting to die.
What happens is that it works great for 10 minutes until the freon and components get soaked with heat and it nudges the pressure just high enough that it shuts off. It will drop pressure and come back on for short cycles and not cool properly.
If that's the case, wait until it quits cooling, then get out and watch the compressor. Normal operation is running almost constantly with the possibility of shutting off ocassionally. Overcharge would be clicking on for a few seconds, then off for several seconds or a minute.
Regarding leaving the gauges on while driving, I don't see a problem with it in theory. Knowing my luck, a hose would fall over and rub on a belt or sit on an exhaust manifold, but as long as the hi and low limit switches are functioning, you can't see pressures that would endanger the gauges.
Might also want to try an experiment. Once it stops cooling, spray the condenser with water. If it suddenly starts working again, likely overcharged.
In reply to AngryCorvair (Forum Supporter) :
Pointofdeparture was my hero in this thread:
https://grassrootsmotorsports.com/forum/grm/volvo-ac-question/184149/page1/
Here's the thread that he suggested and I followed that explains it all in the briefest form:
https://www.matthewsvolvosite.com/ac-shim-fix-are-zip-ties-the-new-bread-clips/
Had I know about this solution, I probably could have saved a ton of money and fixed the failed AC clutch assembly on my old 2005 Tacoma and the PTO clutch on my lawnmower too. I replaced them both a couple of years ago.
Now I know...
Opti
Dork
7/4/21 11:05 p.m.
Id watch pressures, probably give you a better idea of whats happening. If you cant when it stops see if the clutch has power and ground, if it does and its not engaged its a clutch issue. Its the most common failure ive seen with these symptoms. You could also tap on it when it stops and if all the sudden it engages its probably a clutch.
If you suspect a pressure sensor you can just jumper them, or tap on them and sometimes they will start working. Ive also seen evaporators freeze up causing similar symptoms, but rarely is it reliably repeatable.